A new approach for spot detection in total internal reflection fluorescence microscopy

Abstract

Biological images acquired from fluorescence microscopy-based imaging techniques, such as total internal reflection fluorescence microscopy, generally contain hundreds of subcellular components, appearing in the images as bright spots. Therefore, the first step of analysis of these images usually involves the detection of these bright spots. In this paper, we propose a new approach for spot detection using the h-dome transform along with an adaptive mask obtained from regional information. Moreover, local gradient information is used in order to distinguish the spots from other structures. To evaluate the performance of our algorithm, we test it on synthetic images and also real TIRFM images and compare our results with those of two recent methods.